ARLIS/NY Catalogers Discussion Group Meeting
Minutes October 15, 2001
Hosted by the Frick Art Reference Library at the Frick Collection

Attendees: Mark Bresnan, Frick Art Reference Library; Eric Wolf, Frick Art Reference Library; G. Jesse Sadia, Frick Art Reference Library; Kerry Sullivan, Frick Art Reference Library; Rodica L. Preda, Frick Art Reference Library; Amy Schwarz, Frick Art Reference Library; Erin Elliott, Bard Graduate Center; Vicky Bohm, Watson Library; Melanie Wacker, Watson Library; Oleg Kreymer, Watson Library; Annamaria Poma Swank, The Cloisters; Beth Kushner, Brooklyn Museum; Daniel Fermon, MoMA; John Maier, NYU; Sarah Osborne, Morgan Library; Maria Oldal, Morgan Library; Charles Egleston, Morgan Library; Danielle Kovacs, Morgan Library; Gladys Markoff-Sotomayor, Columbia University; Zimra Panitz, Whitney Museum


Mark Bresnan of the Frick Art Reference Library chaired this meeting. After welcoming the group to the Frick Collection he introduced the topics of the meeting: Retrospective Conversion (RECON) and Table of Contents Notes (TOC). These are both projects currently underway at the Frick Art Reference Library (FARL).


Mark Bresnan began the discussion of RECON with a summary and status report of the RECON projects undertaken or underway at the FARL. The first RECON project undertaken here was an in-house project to convert all of the serials holdings from cards to OPAC. This was coordinated by Patricia Siska and is completed, though some holdings information is still being updated from time to time. Rodica Preda was then brought in to coordinate RECON of books; this project was broken into three parts:

  1. Auction Catalogs in the SCIPIO database. Cards were sent to an outside vendor (Electronic Scriptorium) and MARC records were generated from these cards. In-house RECON staff then updated (and continues to update) these records with the actual sale catalogs in hand.
  2. RECON of Monographic Books. This part of FARL RECON project is currently underway. Cards were sent to Duncan Systems Specialists. They will be converted to MARC format, authorized, then brought into OPAC.
  3. RECON of Exhibition Catalogs. Same process as above with additional problems: multiple artists and exhibitions were often cataloged on the same card in card catalog. Should item-level MARC records be generated or should collection-level records be made? The FARL also has a RECON project in its Photoarchive Department. This involves adding to its Artists Authority File the AACR2 or LC forms for artists' names. They are currently reviewing vendor's work in-house on over 24,000 records.
-Questions and Discussion regarding RECON

A discussion of problems libraries have experienced with RECON projects began:

Danny Fermon mentioned the fact that items like CD-ROMs attached to books often slipped through the cracks of RECON project at MoMA. Non-standard records stay that way because library lacks the staff to upgrade all of it records systematically. Vicky Bohm of the Watson mentioned similar problems at her institution. When catalogers find such problems they usually fix them on the record in question, but only this ad-hoc approach is used, not a systematic upgrade of RECON items. Danny Fermon mentioned that MoMA has used the 035 MaRC field (admittedly improperly) to flag RECON records. Beth Kushner then asked about the proper use of the 035 field. Maria Oldal explained that the 035 is a computer-generated field and suggested moving old 035s to 902 field to preserve the information contained there. She further explained that while multiple 035s can be sustained in RLIN, they cannot be in many OPACs including the Voyager system used by the Morgan. Further the 035 is a local field which drops out in RLIN derived records. Danny Fermon reiterated and explained the extra 035s can be added to RLIN records.

Mark Bresnan then asked about RECON of analytics and items without full records. Gladys Markoff of Columbia stated that this information was taken directly from shelflist cards at main library at Columbia, but that the Avery Library went back to the books themselves. In some cases the vendor used records for different editions or completely different titles in deriving records. A project will correct this. Maria Oldal stated that the Morgan RECON was from shelflist cards, though the Reading Room Catalog is far more complete. She stated that this catalog would not be retired anytime soon as it is still very useful to readers and is superior for analytics and records for illuminated manuscripts. There was an effort to do RECON from this catalog, but the process was far too slow - project only got as far as letter "b". Cards are all pre-AACR2 forms. Catalog set up with full record on Main Entry card with other cards as cross-references. Vicky Bohm added that at the Watson the older cards were the better ones.

Maria Oldal explained how some of the cross-references in the old catalog were left out in RECON at the Morgan. Rodica Preda reminded the group that if the library's collection is all in electronic format then the cross-references can be accessed easily by the links in authority records. Vicky Bohm discussed how name changes of museums and AACR2 forms require adding cross-references to the Name Authority File when cataloging older works, particularly catalogs (e.g. Imperial Museum which is now State Museum, etc.).

Rodica Preda explained how her vendor is instructed to use a 955 field to flag non-standard records. Furthermore she discussed how a 6XX with a second indicator of 4 can be used for a subject heading in the correct form which is not established by LC.

She then asked the group which institutions were recently of currently involved in a RECON project and if they had specific policy guidelines outlining these projects. John Maier responded that the Middle-Eastern department at NYU has recently completed a RECON project, but the librarian in charge, Meryle Gaston, will soon depart for Santa Barbara. Maria Oldal then related the RECON policy at the Morgan. Electronic Scriptorium was contracted to create records from the shelflist. The Morgan had provided them with an 80-page instruction manual setting out the guidelines for the project. A sample set of cards was sent to Electronic Scriptorium and reviewed by the Morgan before they sent out the 30-40,000 records from their autograph manuscript collection for RECON. Guidelines were sent telling the vendor to translate pattern "x" on card to pattern "y" in MARC. The Morgan spent a great deal of time preparing the manual. After this, the Morgan did the authority work in-house with a special grant-funded position. She stated that with printed book RECON the vendor's original records were consistently superior to the derives. Rodica Preda agreed that vendor does superb work, but it is essential to provide clear guidelines and properly prepare shelflist cards. Maria Oldal added that it is important to check this work. Each category of record must be checked, important part in quality control process.

Rodica Preda sent out 100 records first in Monograph RECON project. Mark Bresnan asked about multiple artists and venues in RECON projects involving exhibition catalogs. Rodica Preda said effort was made to include all venues and artists when possible. Probably using 600 _4 for artists and 710 or 720 for venues (formerly 797), though this doesn't follow the role of 3. Danny Fermon stated that the rule of 3 is not largely adhered to with venues. Rodica Preda stated a dislike for the 720, but said one cannot use 710 2_ for a non-AACR2 heading. Maria Oldal suggested that one can do whatever one wants if the record is not coded AACR2. Danny Fermon mentioned that his local system did not support 797, etc. Vicky Bohm stated that the Met tags their own exhibits as 600s. Danny Fermon asked for explanation of the difference between 6XX _4 and 690. Maria Oldal explained 6XX _4 follows standard form. Vicky Bohm said that Arms and Armor department uses x90s at Met. Mark Bresnan reminded the group that some libraries like the National Gallery code at 9914 level. John Maier replied that these are the records that people avoid as copy. Maria Oldal reminded the group that RLIN CC code only exists in RLIN and those who catalog locally cannot change this field.



Rodica Preda announced the coming meeting of the New York Technical Services Librarians at the Princeton Club on November 2.

-Table of Contents Notes

Mark Bresnan introduced the topic by stating that the FARL is beginning a project to send recent English language records to a vendor (Blackwell's) who will add 505s to records. He also stated that the FARL ordinarily adds 505s to original records for festschriften and congress proceedings. He then inquired into policies and practices of other libraries. Melanie Wacker and Oleg Kreymer replied that the Met adds contents notes to any book where the Met is involved. Otherwise, contents notes are left to cataloger's judgment. Catalogers provide information and pass item along to para-professionals who do data entry. Annamaria Poma Swank said in some cases a partial contents note can be made. Vicky Bohm said that when artists' names are chapter titles or subtitles the Met adds them separated by $t in 505s in both creates and derives. Maria Oldal asked if people add code to non-enhanced 505s. Vicky Bohm replied that she does. Danny Fermon asked whether one can search only 505s. Maria Oldal answered that one cannot; however, key title word searches achieve largely the same goal. She then asked about sharing records with vendor provided 505s via RLIN. Eric Wolf explained that this could not be accommodated due to intellectual property law; the vendors own the fields they provide. Cataloged records (already in RLIN) are sent to vendor, enhanced, and returned to OPAC, but not uploaded into RLIN.

John Maier wondered if Z3950 protocol could be used to "steal" such enhanced records. Danny Fermon stated that 035s indicate that records have been altered by vendors. Mark Bresnan asked what other libraries use outside vendors for enhancement. Gladys Markoff stated that Columbia uses Blackwell for contents notes. Eric Wolf said he would report more in future meetings when Frick's TOC project was further along.

-Next Meeting

Tentatively set for December 3 at the Bard Graduate Center.
Possible topic: URLs.

Recorded by Eric Wolf